The tumor microenvironment influences both therapeutic outcome and malignant progression. In particular, tumor hypoxia has been shown to be a prognostic indicator for poor therapeutic response, to increase genomic instability, and to induce the expression of genes that increase metastatic spread. Previous studies from my laboratory have indicated that hypoxia induces apoptosis in oncogenically transformed cells in vitro, acts as a selective pressure for the expansion of transformed cells possessing diminished apoptotic potential, and co- localizes with apoptotic regions in tumors. The tumor suppressor protein p53 induces rapid apoptosis in response to oxygen concentrations that induce an S-phase arrest. Hypoxia-induced p53 is nuclear and associates with p53-response elements in target genes, such as mdm2 and p21. Our previous studies have shown that the cellular decision to use p53 transactivation or transrepression is mediated after stress-induced binding of p53 to the promoter, and is determined by the presence of co-activators or co-repressors. In contrast to p53 induced by DNA-damaging agents, hypoxia-induced p53 acts primarily as a repressor. Using extensive microarray analysis, we identified families of repressed targets of p53 that are involved in cell signaling, DNA repair, cell-cycle control and differentiation. A chromatin immunoprecipitation screen demonstrated that p53 induced by either hypoxia or DNA damage binds to the same promoters, regardless of transcriptional outcome. We have also observed that several genes repressed by p53 in normoxia and hypoxia do not appear to have detectable binding by p53 as assayed by chromatin immunoprecipitation. In pursuing this interesting observation, we discovered that at least one of these genes (WNT16B) is regulated through differential histone methylation. This has led to the discovery that the transcriptional repressor JARID1B is a direct p53 target. In a parallel set of studies, we have also determined that HIF is a direct regulator of three Jumonji-domain histone demethylases, including JARID1B. JARID1B represses expression of the proliferative gene CDK6 under hypoxia. The critical hypothesis that we will be testing is whether induction of JARID1B by p53 and/or HIF represses the expression of genes critical for a tumorigenic phenotype. We will test these hypotheses through a combination of genetic and biochemical approaches in cell lines and in mice. By identifying and characterizing the genes repressed by p53 and HIF through JARID1B during hypoxia, our studies should clarify some of the complex regulatory mechanisms activated in the hypoxic microenvironment. PUBLIC HEALTH RELEVANCE: The tumor microenvironment, influences both therapeutic outcome and malignant progression. In particular, tumor hypoxia (reduced oxygen) induces expression of genes that increase the malignant progression of tumors. In hypoxic cancer cells, the tumor suppressor p53 induces cell death by repressing the expression of genes promoting cancer cell survival. We have recently discovered that hypoxia and p53 induce expression of the transcriptional repressor JARID1B, uncovering a novel gene repression pathway. The studies outlined in this proposal will determine the importance of hypoxic repression of JARID1B targets in vitro and in vivo, characterize the importance of p53 in mediating repression under hypoxia, and determine the contribution of JARID1B to tumor progression using spontaneous mouse tumor models.